Thanks for the reply Tom. I figured it must be the loss of the phosphor on the one that went blue. I have no idea what kind of emitter is in those 2AA Defiants….XB-D maybe?
Yes, the center was a little darker than the rest of the hotspot on the other one. How in the heck do you get the emitter farther in the reflector?
Does the S5 reflector have that lip around the base? If so that will bottom out against the pill and thinner insulators won't actually change the focus any. Trim that lip off and it'll start responding to different thickness spacers.
+1. I have some light that needed this too, but still hasn’t done it. Seems way too easy to make a mistake on this and totally ruin the reflector.
What’s a good method to “trim” off the extra base of the reflector?
LOL, I use tin snips. Cuts much easier than you'd think. Needs only very little cleanup with a file afterwards to knock off the sharp edge, the whole thing doesn't have to be filed down perfectly smooth.
Dremel?
I tried to dedome a dead xm-l in gas for about 16h, but it failed. I hanged the star sideways in the glass, i thought the dome will slip off by the gravity. It actually worked, but a lot of silicone remained near the wires. I tried to remove, but i broke the wires.
Is there a way to trick out that situation? Dissolve the silicone, or anything? Because i think if i do the same thing again it will also broke, and it is a bit pricey to waste the xm-ls to me.
Oh boy - I got a ton of XML T6's laying around, all from upgrade mods. I always do it on a dis-connected board though, never wired. I use a toothpick now with a 20x lighted magnifier to clean up the silicone, but I won't try for 100% around the wires. You have to use something soft with a magnifier, even then, I don't take the risk getting too close to the wires. Very important though to have good visibility.
For “wires” i meant those hair-thin copper wires on the led die. Mine is also disconnected. The problem was that a silicone chunk that remained on the die with the copper wires in it were loose, so i thought it released the wores, i could simply push it away from the led.
My first XML dedoming in progress (XPG was a fail 
), will see how it turn out…
I burned up an XM-L (& driver) while bench testing a driver recently (thanks to RaceR86! - just kidding man) and so I was about to toss it in the trash when I had this idea to try and dedome it by soaking it in gas. I had not brushed up on the technical details on what to do and didn't have time to so I just threw it in a plastic cap with some gas and set it aside (cap was about 1 inch wide and 1 inch tall and I only put about 1/4 to 1/2 inch of gas in it). I checked on it every now and then and finally plucked it out about 2 hours later. The dome just fell off and it looked fairly clean. I then doused it with some rubbing alcohol to rinse it. I did try applying power to it to see if the gasoline brought it back to life, but no such luck 
. I'm sure if this LED had been a working one I would have failed.
Here's a pic (best I could get with my phone):
-Garry
Mine ended up kinda OK-ish, but tint shifted to more-green :Sp which I don’t like too much…
Do we have a guide as to how far tint shifts based on the tint you start with before dedoming? I know generally speaking that dedoming will move you to a warmer tint, but how far? What is the "best" cool white tint to start with so as to end up near a nice neutral white tint (i.e. 4C tint)? Start with a 1C? 1A? Or is it not this simple?
-Garry
I saw a nice post of maybe 3 different tints, and their resulting de-domed tints - not sure where this was. I like the 0D and 1A's de-domed - 1A turns neutral but not too much yellow - much like a 3C or little less. 1C's definite more yellow, I'd say more so than a 3C. I accidentally de-domed a 3C and wow - ton of yellow but looked nothing like the T3 6A1 and T4 warms I've seen - the de-domed 3C was much more yellow than the warm tints, in fact the warm tints seem to me to be more of a true pale white while the 3C is more yellow.
Well thanks to RaceR86 i know that de doming of cold white emitter(6500) upper neutral white color appears(around 4800) which has little greenish tint visible in aspheric and not so visible in reflector. This de-domed tint looks much better to me than cool white but don’t ask me; you decide.
Guys if you are not familiar with tint colors it looks something like this:
I would say its beneficial to start with 1A/1D instead of 1B/C.
On the tint chart, most emitters seem to end up 1300-2100K to the left, and then some up (towards the green/yellow spectrum) You will not end up with anything as nice as a 4C no matter where you start in the CW range.. A 2A would probably get you "close" to a 4C in terms of temperature. But it could still not compete in color rendition or color tone IMO. I have not measured this stuff, so I dont have hard factual numbers. I have seen texaspyro show some stuff on temperature change, but I have never seen any numbers on color tone.
Luminarium, your are learning. :) Although the pictures you are showing are for describing temperature (kelvin), not hue.
I’ve found through trial and ONE error (XR-E) that dedoming an XP-G2 R5 2B (CW) on a noctigon:
http://intl-outdoor.com/noctigon-xp20-mcpcb-cree-xpg2-r5-2b-led-p-723.html
:and put in a DD zoomie gave me a really sharp die image, and it retained a great amount of ‘cool whiteness’…
But, when I dedomed another XP-G2 R5 2B (CW), this time on a alu star:
http://intl-outdoor.com/cree-xpg2-r5-2b-led-20mm-mcpcb-p-705.html
:and again mounted in an identical DD zoomie, I got the green tint… everso slight on zoomed in mode (again a nice sharp die image), but quite visible on flood!
Clearly the Noctigon copper base helps immensely.
Cheers all
Rob
RaceR86 - that's what I was thinking - de-doming would take you 3-4 cells over to the right, and maybe 1 up on this chart: http://flashlightwiki.com/images/c/cb/Ansi-white.jpg. Don't think your link is working, not sure if the ANSI white chart is what you meant.
Link works for me.
-Garry
Yesterday tried my first gasoline dedome, in EU unleaded 95 gas (here called "benzina", if anybody cares).
Worked like a dream, both emitters came out of the overnight bath clean like the dome was never been there, wires area too! GREAT!
But clearly I underestimated the color shift: on noctigon XML2 T6 C3 became so yellow it looks like those french car headlights from the early eighties... 
And on octigon XPG2 R5 B2 went slightly better but has a definite green hue.
Now ordered XML2 U2 A1, hope the tint remain better...
Did my first dedoming for a TR-J19 build. Soaked the emitters in gasoline overnight and cleaned them up this morning. Tested them and they all work. :)
